mmdetection-2.26.0/configs/lad/lad_r50_paa_r101_fpn_coco_1x.py

_base_ = [
    '../_base_/datasets/coco_detection.py',
    '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py'
]
teacher_ckpt = 'http://download.openmmlab.com/mmdetection/v2.0/paa/paa_r101_fpn_1x_coco/paa_r101_fpn_1x_coco_20200821-0a1825a4.pth'  # noqa
model = dict(
    type='LAD',
    # student
    backbone=dict(
        type='ResNet',
        depth=50,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        norm_cfg=dict(type='BN', requires_grad=True),
        norm_eval=True,
        style='pytorch',
        init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50')),
    neck=dict(
        type='FPN',
        in_channels=[256, 512, 1024, 2048],
        out_channels=256,
        start_level=1,
        add_extra_convs='on_output',
        num_outs=5),
    bbox_head=dict(
        type='LADHead',
        reg_decoded_bbox=True,
        score_voting=True,
        topk=9,
        num_classes=80,
        in_channels=256,
        stacked_convs=4,
        feat_channels=256,
        anchor_generator=dict(
            type='AnchorGenerator',
            ratios=[1.0],
            octave_base_scale=8,
            scales_per_octave=1,
            strides=[8, 16, 32, 64, 128]),
        bbox_coder=dict(
            type='DeltaXYWHBBoxCoder',
            target_means=[.0, .0, .0, .0],
            target_stds=[0.1, 0.1, 0.2, 0.2]),
        loss_cls=dict(
            type='FocalLoss',
            use_sigmoid=True,
            gamma=2.0,
            alpha=0.25,
            loss_weight=1.0),
        loss_bbox=dict(type='GIoULoss', loss_weight=1.3),
        loss_centerness=dict(
            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=0.5)),
    # teacher
    teacher_ckpt=teacher_ckpt,
    teacher_backbone=dict(
        type='ResNet',
        depth=101,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        norm_cfg=dict(type='BN', requires_grad=True),
        norm_eval=True,
        style='pytorch'),
    teacher_neck=dict(
        type='FPN',
        in_channels=[256, 512, 1024, 2048],
        out_channels=256,
        start_level=1,
        add_extra_convs='on_output',
        num_outs=5),
    teacher_bbox_head=dict(
        type='LADHead',
        reg_decoded_bbox=True,
        score_voting=True,
        topk=9,
        num_classes=80,
        in_channels=256,
        stacked_convs=4,
        feat_channels=256,
        anchor_generator=dict(
            type='AnchorGenerator',
            ratios=[1.0],
            octave_base_scale=8,
            scales_per_octave=1,
            strides=[8, 16, 32, 64, 128]),
        bbox_coder=dict(
            type='DeltaXYWHBBoxCoder',
            target_means=[.0, .0, .0, .0],
            target_stds=[0.1, 0.1, 0.2, 0.2]),
        loss_cls=dict(
            type='FocalLoss',
            use_sigmoid=True,
            gamma=2.0,
            alpha=0.25,
            loss_weight=1.0),
        loss_bbox=dict(type='GIoULoss', loss_weight=1.3),
        loss_centerness=dict(
            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=0.5)),
    # training and testing settings
    train_cfg=dict(
        assigner=dict(
            type='MaxIoUAssigner',
            pos_iou_thr=0.1,
            neg_iou_thr=0.1,
            min_pos_iou=0,
            ignore_iof_thr=-1),
        allowed_border=-1,
        pos_weight=-1,
        debug=False),
    test_cfg=dict(
        nms_pre=1000,
        min_bbox_size=0,
        score_thr=0.05,
        score_voting=True,
        nms=dict(type='nms', iou_threshold=0.6),
        max_per_img=100))
data = dict(samples_per_gpu=8, workers_per_gpu=4)
optimizer = dict(lr=0.01)
fp16 = dict(loss_scale=512.)

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# USER SHOULD NOT CHANGE ITS VALUES.
# base_batch_size = (8 GPUs) x (8 samples per GPU)
auto_scale_lr = dict(base_batch_size=64)